The prevalence of insulin resistance in glucose intolerant subjects has long been recognized. Reaven et al (American Journal of Medicine 1976, 60, 80) used a continuous infusion of glucose and insulin (insulin/glucose clamp technique) and oral glucose tolerance tests to demonstrate that insulin resistance existed in a diverse group of nonobese, nonketotic subjects. These subjects ranged from borderline glucose tolerant to overt, fasting hyperglycemia. The diabetic groups in these studies included both insulin dependent (IDDM) and noninsulin dependent (NIDDM) subjects.
Coincident with sustained insulin resistance is the more easily determined hyperinsulinemia, which can be measured by accurate determination of circulating plasma insulin concentration in the plasma of subjects. Hyperinsulinemia can be present as a result of insulin resistance, such as is in obese and/or diabetic (NIDDM) subjects and/or glucose intolerant subjects, or in IDDM subjects, as a consequence of over injection of insulin compared with normal physiological release of the hormone by the endocrine pancreas.
The association of hyperinsulinemia with obesity and with ischemic diseases of the large blood vessels (e.g. atherosclerosis) has been well established by numerous experimental, clinical and epidemiological studies (summarized by Stout, Metabolism 1985, 34, 7, and in more detail by Pyorala et al, Diabetes/Metabolism Reviews 1987, 3, 463). Statistically significant plasma insulin elevations at 1 and 2 hours after oral glucose load correlates with an increased risk of coronary heart disease.
Since most of these studies actually excluded diabetic subjects, data relating the risk of atherosclerotic diseases to the diabetic condition are not as numerous, but point in the same direction as for nondiabetic subjects (Pyorala et al). However, the incidence of atherosclerotic diseases in morbidity and mortality statistics in the diabetic population exceeds that of the nondiabetic population (Pyorala et al; Jarrett Diabetes/Metabolism Reviews 1989, 5, 547; Harris et al, Mortality from diabetes, in Diabetes in America 1985).
The independent risk factors obesity and hypertension for atherosclerotic diseases are also associated with insulin resistance. Using a combination of insulin/glucose clamps, tracer glucose infusion and indirect calorimetry, it has been demonstrated that the insulin resistance of essential hypertension is located in peripheral tissues (principally muscle) and correlates directly with the severity of hypertension (DeFronzo and Ferrannini, Diabetes Care 1991, 14, 173). In hypertension of the obese, insulin resistance generates hyperinsulinemia, which is recruited as a mechanism to limit further weight gain via thermogenesis, but insulin also increases renal sodium reabsorption and stimulates the sympathetic nervous system in kidneys, heart, and vasculature, creating hypertension.
It is now appreciated that insulin resistance is usually the result of a defect in the insulin receptor signaling system, at a site post binding of insulin to the receptor. Accumulated scientific evidence demonstrating insulin resistance in the major tissues which respond to insulin (muscle, liver, adipose), strongly suggests that a defect in insulin signal transduction resides at an early step in this cascade, specifically at the insulin receptor kinase activity, which appears to be diminished (reviewed by Haring, Diabetalogia 1991, 34, 848).
Protein-tyrosine phosphatases (PTPases) play an important role in the regulation of phosphorylation of proteins. The interaction of insulin with its receptor leads to phosphorylation of certain tyrosine molecules within the receptor protein, thus activating the receptor kinase. PTPases dephosphorylate the activated insulin receptor, attenuating the tyrosine kinase activity. PTPases can also modulate post-receptor signaling by catalyzing the dephosphorylation of cellular substrates of the insulin receptor kinase. The enzymes that appear most likely to closely associate with the insulin receptor and therefore, most likely to regulate the insulin receptor kinase activity, include PTP1B, LAR, PTPα and SH-PTP2 (B. J. Goldstein, J. Cellular Biochemistry 1992, 48, 33; B. J. Goldstein, Receptor 1993, 3, 1–15; F. Ahmad and B. J. Goldstein Biochim. Biophys Acta 1995, 1248, 57–69).
McGuire et al. (Diabetes 1991, 40, 939), demonstrated that nondiabetic glucose intolerant subjects possessed significantly elevated levels of PTPase activity in muscle tissue vs. normal subjects, and that insulin infusion failed to suppress PTPase activity as it did in insulin sensitive subjects.
Meyerovitch et al (J. Clinical Invest. 1989, 84, 976) observed significantly increased PTPase activity in the livers of two rodent models of IDDM, the genetically diabetic BB rat, and the STZ-induced diabetic rat. Sredy et al (Metabolism, 44, 1074, 1995) observed similar increased PTPase activity in the livers of obese, diabetic ob/ob mice, a genetic rodent model of NIDDM.
The compounds of this invention have been shown to inhibit PTPases derived from rat liver microsomes and human-derived recombinant PTPase-1B (hPTP-1B) in vitro. They are useful in the treatment of insulin resistance associated with obesity, glucose intolerance, diabetes mellitus, hypertension and ischemic diseases of the large and small blood vessels.
P. N. Devine et al (WO 97/21693; Jun. 19, 1997) disclosed examples D under a method of preparation (B, D (independently=halogen, phenyl, alkyl; X=alkyl, aryl; Y=(CH2)0-3CH3, Ph, NH(CH2)0-3CH3, N((CH2)0-3CH3)2, NH2, NO2, NHCO(CH2)0-3CH3, NHCO2(CH2)0-3CH3, CH2O(CH2)0-3CH3, OPh; O(CH2)1-4O(CH2)0-5CH3, O(CH2)1-4OPh, OCO2(CH2)0-5CH3, CON((CH2)0-5CH3)2, O(CH2)1-4O(CH2)1-6Ph). The synthetic process to prepare the compounds represented by compounds D was different to the processes used to prepare the 2,3,5-substituted biphenyls of this invention. 
G. Cain and C. J. Eyermann (U.S. Pat. No. 5,523,302; Jun. 4, 1996) disclosed examples A (B, D (independently=cycloalkyl, alkyl, aralkyl) as agents which inhibit platelet aggregation, as thrombolytics, and/or for the treatment of thromboembolic disorders. The synthetic process to prepare the compounds represented by compounds A was different to the processes used to prepare the 2,3,5-substituted biphenyls of this invention. 
M. Wayne et al (WO 94/22835, WO 94/22834; Oct. 13, 1994) disclosed examples B (B, D (independently=alkyl, halogen) as agents which inhibit platelet aggregation, as thrombolytics and/or for the treatment of thromboembolic disorders. The synthetic process to prepare the compounds represented by compounds B was different to the processes used to prepare the 2,3,5-substituted biphenyls of this invention 
S. W. Bagley et al (U.S. Pat. No. 5,334,598; Aug. 2, 1994) disclosed examples C (B, D (independently=phenyl, naphthyl, alkyl, halogen) as agents which have endothelin antagonist activity and are therefore useful in treating cardiovascular disorders. Our present invention does not claim a compound of this genus, namely 2-phenyl-2-phenoxy acetic acids. The synthetic process to prepare the compounds represented by compounds C was different to the processes used to prepare the 2,3,5-substituted biphenyls of this invention. A similar set of compounds is disclosed in C. M. Harvey et al (WO 96/09818; Apr. 4, 1996), W. J. Greenlee et al (WO 91/11909; Aug. 22, 1991), and W. J. Greenlee et al (WO 91/12002; Aug. 22, 1991). A similar set of arguments apply. 